Date: Mon, 27 Apr 2026 18:00:37 -0400 From: Mark Johnston <markj@freebsd.org> To: Andrew Turner <andrew@freebsd.org> Cc: src-committers@freebsd.org, dev-commits-src-all@freebsd.org, dev-commits-src-main@freebsd.org, Sarah Walker <sarah.walker2@arm.com> Subject: Re: git: c16c95192f01 - main - virtio_net: Use bus_dma for rxq/txq buffers Message-ID: <ae_chW7sJV1YAa-h@nuc> In-Reply-To: <69ef4acf.1fcd2.4d4d3e54@gitrepo.freebsd.org>
index | next in thread | previous in thread | raw e-mail
On Mon, Apr 27, 2026 at 11:38:55AM +0000, Andrew Turner wrote: > The branch main has been updated by andrew: > > URL: https://cgit.FreeBSD.org/src/commit/?id=c16c95192f01237a876eb7bc336e3bbda9310171 > > commit c16c95192f01237a876eb7bc336e3bbda9310171 > Author: Sarah Walker <sarah.walker2@arm.com> > AuthorDate: 2026-02-16 14:19:13 +0000 > Commit: Andrew Turner <andrew@FreeBSD.org> > CommitDate: 2026-04-27 11:37:53 +0000 > > virtio_net: Use bus_dma for rxq/txq buffers > > While the majority of virtio platforms will be fully coherent, some may > require cache maintenance or other specific device memory handling (eg for > secure partitioning). Using bus_dma allows for these usecases. > > The virtio buffers are marked as coherent; this should ensure that sync > calls are no-ops in the common cases. > > Reviewed by: andrew > Sponsored by: Arm Ltd > Differential Revision: https://reviews.freebsd.org/D55492 After this commit I see a ton of warnings during boot in an amd64 bhyve VM: uma_zalloc_debug: zone "malloc-16" with the following non-sleepable locks held: exclusive sleep mutex vtnet0-rx0 (vtnet0-rx0) r = 0 (0xfffffe00f1e71d80) locked @ /home/markj/sb/main/src/sys/dev/virtio/network/if_vtnet.c:3584 exclusive sleep mutex vtnet0 (VTNET Core Lock) r = 0 (0xfffffe00f1b819a8) locked @ /home/markj/sb/main/src/sys/dev/virtio/network/if_vtnet.c:1572 stack backtrace: #0 0xffffffff8134b497 at witness_debugger+0x137 #1 0xffffffff8134dad9 at witness_warn+0x749 #2 0xffffffff819524bc at uma_zalloc_debug+0xbc #3 0xffffffff81951df0 at uma_zalloc_arg+0x30 #4 0xffffffff812083cb at malloc+0xab #5 0xffffffff8126f92c at sysctl_add_oid+0x7c #6 0xffffffff81b92fcd at alloc_bounce_zone+0x58d #7 0xffffffff81b91164 at bounce_bus_dmamap_create+0x294 #8 0xffffffff80ededc6 at vtnet_rx_alloc_buf+0x166 #9 0xffffffff80ee2ce2 at vtnet_init_locked+0xb42 #10 0xffffffff80ee10da at vtnet_ioctl+0x33a #11 0xffffffff814b63da at ifhwioctl+0x153a #12 0xffffffff814b81bb at ifioctl+0x76b #13 0xffffffff81356d36 at kern_ioctl+0x406 #14 0xffffffff8135687c at sys_ioctl+0x26c #15 0xffffffff81bffea5 at amd64_syscall+0x3d5 #16 0xffffffff81bb77fb at fast_syscall_common+0xf8 It seems surprising that we need to create a UMA zone at all.home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ae_chW7sJV1YAa-h>